Y Balance Test (YBT) Dataset

The following is the dataset experimented in the paper "Scoring Performance on the Y-Balance Test" [6] by Mahato et al., presented at 27th International Conference on Case-Based Reasoning.

YBT Test

The Y Balance Test (YBT) is the most common dynamic balance assessment used within the sports medicine clinical context[2]. It requires an individual to transition from a position of bilateral to unilateral stance, perform a maximal reach excursion with the non-stance limb in three standardised directions (anterior; posteromedial; posterolateral), while maintaining controlled balance[2] (see Figure 1).

Figure 1: A demonstration of the YBT test in operation.

The individual is then required to return to the starting position in a controlled manner. A trial is deemed a fail if they remove their hands from their hips, make contact with the ground, weight bear through the slider, raise the stance leg heel or kick the slider forward for extra distance. Participants typically complete four practice trials prior to completion of three recorded trials in each direction (randomized order), bilaterally[2]. The traditional balance score is obtained by manually measuring the distance the individual reaches outside of their base of support and normalising it to their leg length, allowing for appropriate comparison between individuals. Previous research has demonstrated the ability of this protocol to identify differences in dynamic performance between control and pathological groups, in conditions such as acute lateral ankle sprain[1] and anterior cruciate ligament injuries[3]. It has also been suggested that the YBT may have a role in evaluating concussed athletes. It can provide a means to challenge the sensorimotor subsystems of injured athletes, highlighting deficits that may increase their risk of sustaining further injury[5]. Johnston et al. have shown that a very good assessment of YBT performancecan be obtained from a single wearable sensor. [4] Normal and abnormal balance performance can be assessed with a moderate level of accuracy.

Data Desciption

The data set consists of data collected from 29 young healthy adults (aged 23.3 ∓ 2.1 years; height 174.7 ∓ 9.2 cm; weight 71.6 ∓ 13.3 kg; left leg length 95.4 ∓ 4.8 cm; right leg length 95.5 ∓ 5.1 cm) were tested on one occasion in a university biomechanics laboratory.

Each sample of data contains the following attributes:

1. age: The age of the subject.
2. cohort: The cohort the subject belongs to (in this dataset all belong to the same cohort).
3. data: A pandas dataframe of data collected by the Shimmer3 sensor (90 features).
4. direction: The direction the subject was trying to reach (0 = anterior; 1 = posteromedial; 2 = posterolateral).
5. dominant: The dominant leg of the subject (0 = left; 1 = right).
6. height: The height of the subject.
7. left_leg: The length of the left leg of the subject.
8. norm_reach: The normalised reach (raw reach / length of the leg) distance of the subject on performing a successful rep.
9. raw_reach: The raw reach distance of the subject on performing a successful rep.
10. right_leg: The length of the right leg of the subject.
11. stance: The leg on which the subject was standing while performing the rep (0 = left; 1 = right).
12. subject: The subject's unique ID.
13. trial: The trial number of the experiment (1-3).

Data Downloads

• YBT_data.npy : The dataset - a Python dict dumped as Numpy object (161MB).
• TSReach.py : A class file that is required to access the data attributes (should be present in the same class).
• Loading Data.ipynb : A demo python jupyter notebook, that shows how to load and access the data.

Contact Author

References

1. Doherty, C., Bleakley, C.M., Hertel, J., Caulfield, B., Ryan, J., Delahunt, E.: Laboratory measures of postural control during the star excursion balance test after acute first-time lateral ankle sprain. Journal of athletic training 50(6), 651-664 (2015)
2. Gribble, P.A., Hertel, J., Plisky, P.: Using the star excursion balance test to assess dynamic postural-control deficits and outcomes in lower extremity injury: a literature and systematic review. Journal of athletic training 47(3), 339-357 (2012).
3. Herrington, L., Hatcher, J., Hatcher, A., McNicholas, M.: A comparison of star excursion balance test reach distances between acl deficient patients and asymptomatic controls. The Knee 16(2), 149-152 (2009).
4. Johnston, W., O'Reilly, M., Dolan, K., Reid, N., Coughlan, G., Caulfield, B.: Objective classification of dynamic balance using a single wearable sensor. In: 4th International Congress on Sport Sciences Research and Technology Support 2016, Porto, Portugal, 7-9 November 2016. pp. 15-24. SCITEPRESS{Science and Technology Publications (2016).
5. Johnston, W., O'Reilly, M., Argent, R., Caulfield, B.: Reliability, validity and utility of inertial sensor systems for postural control assessment in sport science and medicine applications: A systematic review. Sports Medicine pp. 1-36 (2019).
6. Mahato, V., Johnston, W., Cunningham, P.: Scoring performance on the y-balancetest. 27th International Conference on Case-Based Reasoning (2019).